Documentos de Académico
Documentos de Profesional
Documentos de Cultura
Abstract
Background: Bone and joint infection in adults arises most commonly as a complication of joint replacement
surgery, fracture fixation and diabetic foot infection. The associated morbidity can be devastating to patients and
costs the National Health Service an estimated £20,000 to £40,000 per patient.
Current standard of care in most UK centres includes a prolonged course (4–6 weeks) of intravenous antibiotics
supported, if available, by an outpatient parenteral antibiotic therapy service. Intravenous therapy carries with it
substantial risks and inconvenience to patients, and the antibiotic-related costs are approximately ten times that of
oral therapy. Despite this, there is no evidence to suggest that oral therapy results in inferior outcomes.
We hypothesise that, by selecting oral agents with high bioavailability, good tissue penetration and activity against
the known or likely pathogens, key outcomes in patients managed primarily with oral therapy are non-inferior to
those in patients treated by intravenous therapy.
Methods: The OVIVA trial is a parallel group, randomised (1:1), un-blinded, non-inferiority trial conducted in thirty
hospitals across the UK. Eligible participants are adults (>18 years) with a clinical syndrome consistent with a bone,
joint or metalware-associated infection who have received ≤7 days of intravenous antibiotic therapy from the date
of definitive surgery (or the start of planned curative therapy in patients treated without surgical intervention).
Participants are randomised to receive either oral or intravenous antibiotics, selected by a specialist infection
physician, for the first 6 weeks of therapy. The primary outcome measure is definite treatment failure within one
year of randomisation, as assessed by a blinded endpoint committee, according to pre-defined microbiological,
histological and clinical criteria. Enrolling 1,050 subjects will provide 90 % power to demonstrate non-inferiority,
defined as less than 7.5 % absolute increase in treatment failure rate in patients randomised to oral therapy as
compared to intravenous therapy (one-sided alpha of 0.05).
Discussion: If our results demonstrate non-inferiority of orally administered antibiotic therapy, this trial is likely to
facilitate a dramatically improved patient experience and alleviate a substantial financial burden on healthcare
services.
Trial registration: ISRCTN91566927 - 14/02/2013
Keywords: Bone and joint infection, Oral, Intravenous, Antibiotic, Non-inferiority, Treatment failure
* Correspondence: ho-kwong.li@ouh.nhs.uk
1
Nuffield Department of Medicine, University of Oxford, Oxford, UK
Full list of author information is available at the end of the article
© 2015 Li et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International
License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any
medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://
creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Li et al. Trials (2015) 16:583 Page 2 of 12
clinical practice (GCP) trained research staff after asses- 4) Mild osteomyelitis, defined as bone infection which,
sing their understanding of the Patient Information in the opinion of the clinical investigator, would not
Sheet (PIS) [see Additional files 1 and 2]. Eligibility is usually require a 6-week course of intravenous
determined by the inclusion and exclusion criteria listed antibiotic therapy
below. 5) An infection for which there are no suitable
antibiotic choices to permit randomisation
Inclusion criteria between the two arms of the trial (for example,
The participant may be enrolled only if he or she meets where organisms are only sensitive to intravenous
each of the following criteria: antibiotics)
6) Previous enrolment in the trial
1) Has a clinical syndrome comprising any of the 7) Septic shock or systemic features requiring
following: a) localised pain or b) localised erythema intravenous antibiotic therapy in the opinion of
or c) temperature >38.0 °C or d) a discharging sinus the treating clinician (the patient may be re-evaluated
or wound if these features resolve)
2) Is willing and able to give informed consent 8) Evidence of being unlikely to comply with trial
3) Is aged 18 years or above requirements following randomisation in the opinion
4) Has received 7 days or less of intravenous therapy of the investigator
after an appropriate surgical intervention to treat 9) Clinical, histological or microbiological evidence of
bone or joint infection (regardless of pre-surgical mycobacterial, fungal, parasitic or viral aetiology of
antibiotics) or, if no surgical intervention is required, the infection
the patient has received 7 days or less of intravenous 10) Receiving an investigational medical product as part
therapy after the start of treatment for the relevant of another clinical trial
clinical episode
5) Has a life expectancy > 1 year The use of antibiotic-loaded cement in spacers, bone
6) Has a bone and joint infection in one of the following substitutes or beads at the site of infection is not an ex-
categories: a) native osteomyelitis (bone infection clusion criterion, but is recorded at baseline. Pregnancy,
without metalwork) including haematogenous or renal failure and liver failure are not exclusion criteria,
contiguous osteomyelitis; b) native joint sepsis treated provided suitable antibiotic options can be identified for
by excision arthroplasty; c) prosthetic joint infection both IV and PO therapy prior to randomisation.
treated by debridement and retention, by one stage
revision or by excision of the prosthetic joint (with or Randomisation
without planned re-implantation); d) orthopaedic An electronic randomisation service, with telephone
device or bone-graft infection treated by debridement backup if necessary, is provided through a clinical trials
and retention or by debridement and removal; unit. After confirming the patient’s eligibility, the ran-
e) spinal infection, including discitis, osteomyelitis domisation service assigns a sequentially allocated study
or epidural abscess. number and informs the investigator of the treatment
allocation in real time and by confirmatory e-mail. Ran-
Exclusion criteria domisation is stratified by site, to take account of vari-
The participant may not enter the study if he or she has ation in clinical practice among centres.
any one of the following: The local clinician or study nurse is responsible for
documenting the participant’s enrolment in their clinical
1) Staphylococcus aureus bacteraemia on presentation notes and for informing the participant’s general
or within the last 1 month practitioner.
2) Bacterial endocarditis, either on presentation or
within the last month (NB: there are no study Minimising bias
mandated investigations, so participants are not Blinding is not possible, since we consider that giving
required to have echocardiograms, blood cultures, prolonged intravenous placebo would pose an unneces-
or any other investigations to exclude endocarditis sary risk to participants and therefore would be uneth-
in the absence of a clinical indication) ical. Because an open label study is at risk of bias, we
3) Any other concomitant infection which, in the have appointed an endpoint review committee to review
opinion of the clinician responsible for the patient, the clinical notes relating to any patient who fulfils the
requires a prolonged course of intravenous antibiotic criteria for a potential treatment failure. All notes
therapy (for example, mediastinal infection or reviewed are redacted for personal identifiable data and
central nervous system infection) for any information that might indicate the choice and
Li et al. Trials (2015) 16:583 Page 4 of 12
route of antibiotic therapy. The endpoint review com- changing from one treatment strategy to the other, since
mittee is therefore blind to treatment allocation. The there is equipoise regarding efficacy.
notes are examined against objective criteria for meeting There are no formal withdrawal criteria in this study
the primary endpoint. other than at the request of a participant. All patients
are free to withdraw their consent at any time; if they
Study interventions: PO versus IV antibiotic strategy elect to withdraw from the allocated treatment strategy
The selection of individual antibiotic agents within the during the randomised treatment phase, they will meet a
allocated strategy (that is, PO or IV antibiotics) is the re- secondary endpoint but may still be followed up and be
sponsibility of the infection specialist caring for the pa- included in the analysis provided that appropriate con-
tient; the choice depends on microbiological assessment, sent is given.
the side effect profile and patient preference. In the Any medical decision to withdraw a participant from
event of a culture-negative bone or joint infection (or the randomised strategy is discussed with the chief in-
where there is a delay in availability of culture results), vestigator (CI) or the trial physician. Changing the anti-
the infection specialist selects the most appropriate em- biotic whilst remaining within the allocated strategy
piric therapy. If new information becomes available, the need not be discussed, but a clinician with appropriate
infection specialist may change the choice of antibiotic training in managing infection must make such a
agent but should remain within the allocated route of decision.
administration strategy; if this is not possible, the partici- The infection specialist may adjust dosage based on
pant will have reached an endpoint. renal or hepatic function, drug interactions or other fac-
If randomised to IV strategy, participants are expected tors according to drug labelling information, the British
to complete 6 weeks of IV antibiotic therapy. Where it is National Formulary and local pharmacy guidelines. All
common practice to use adjunctive oral agents in antibiotics used (including dosage, route of adminis-
patients treated with IV therapy (for example, oral rifam- tration and duration) are recorded from the day of
picin as adjunctive therapy for biofilm-related staphylo- randomisation through to final follow-up at one year
coccal infection), such therapy will be allowed. (Fig. 1).
If randomised to the PO strategy, participants are
expected to commence PO therapy within seven days Assessments
of definitive surgical management (or, if managed When a participant is an inpatient, the study clinician or
without surgery, from the start of antibiotic therapy research nurses will maintain contact with the clinical
for that clinical episode). If a participant allocated to team to identify potential endpoints, and to ensure
the PO strategy requires IV antibiotic therapy for an implementation of the randomised antibiotic strategy.
unrelated intercurrent illness during the initial 6 weeks Following hospital discharge, participants are seen ac-
of treatment, or experiences vomiting or inability to cording to clinically determined follow-up plans. Trial-
swallow, or concerns arise about absorption of oral specific data are obtained from the case record at
medication, then IV antibiotic therapy may be substituted 6 weeks (range from day 21 to day 63), 4 months (range
for up to five days. If IV antibiotic prescribing ex- day 70 to day 180) and 1 year (range day 250 to day
ceeds five days, the patient will have met a secondary 420). If the patient does not attend clinic within the spe-
endpoint, but will still contribute to the intention-to- cified date ranges, the investigator arranges a telephone
treat analysis. review with the participant or the participant’s general
Follow-on antibiotic treatment after the initial 6 weeks practitioner (GP) to identify potential endpoints or ser-
may be prescribed in either arm of the trial, but the ious adverse events. If, based on the telephone discus-
choice of agent, duration and route of administration are sion, clinical review is indicated, the investigator
not governed by the trial protocol. organises this and advises the patient accordingly.
If continuing in the randomised strategy (IV or PO) is A study clinician reviews the source documents from
no longer compatible with optimal clinical care, the clin- routine care visits when completing investigator reviews.
ician must commence appropriate alternative therapy. They record:
Appropriate reasons for discontinuing the allocated
treatment might include lack of suitable medication a) Microbiology and histology results and date of
within the allocated strategy, adverse reactions, contrain- discharge (first review only)
dications or susceptibility testing results. Failure to b) Outpatient visits since randomisation
maintain intravenous access is an appropriate reason for c) Serious adverse events
discontinuing IV antibiotics. A wound discharge or other d) Re-admissions for inpatient care
clinical signs related to the incident infection, or its e) Type of intravenous catheter (line) used and any
apparent resolution, are not appropriate indications for line-related complications
Li et al. Trials (2015) 16:583 Page 5 of 12
Eligible for trial and has completed seven days or less of IV treatment
Informed Consent
Randomise
Monitor
Further antibiotics if progress, but
clinically indicated antibiotic choice
(not part of study) not influenced
by study.
f ) Episodes of Clostridium difficile associated diarrhoea 2. IV line complications (infection, thrombosis or other
g) Antibiotic use to date (including dosage, route and events requiring early removal or replacement of
model of care, for example, district nurse, self- the line)
administered or daily clinic visits) 3. “Probable” or “possible” treatment failure, as
h) Presence or absence of any potential endpoints composites with definite treatment failure. These
i) Reasons for not completing the planned antibiotic are determined by blinded endpoint committee
course (if applicable). review and defined by the presence of any one of
the following criteria:
Outcome measures
Endpoints are identified by prospective surveillance a) Loosening of a prosthesis, confirmed radiologically
throughout the year post-randomisation. b) Non-union of a fracture after 6 months, confirmed
The primary endpoint is definite failure of infection radiologically
treatment, where definite failure is indicated by one or c) Superficial spreading erythema, treated as cellulitis
more of the following: with an antibiotic for >1 week; where results from
deep tissue samples do not meet the primary
Isolating bacteria from two or more samples of endpoint as described above
bone/spine/peri-prosthetic tissue, where the bacteria
are phenotypically indistinguishable Where appropriate deep tissue samples have been sent
Isolating a pathogenic organism (for example, for microbiology and results of the culture are negative,
Staphylococcus aureus, but not Staphylococcus and any of a), b) or c) are met, then the endpoint will be
epidermidis) on a single, closed biopsy of native regarded as “possible”. On the other hand, where deep
bone or spine tissue samples are not sent for microbiology, and any of
Diagnostic histologic findings on bone or peri- a), b) or c) are met, then the endpoint will be regarded
prosthetic tissue as “probable”
Formation of a draining sinus tract arising from
bone or prosthesis 4. Early termination of the planned 6-week period
Recurrence of frank pus adjacent to bone or of oral or IV antibiotic therapy because of
prosthesis. adverse events, patient preference or any other
reason
Secondary endpoints are: 5. Resource allocation determined by: a) length of
inpatient hospital stay; b) frequency of outpatient
1. Serious adverse events (SAEs), including death (that visits; or c) antibiotic prescribing costs
is, all-cause mortality) 6. Quality of life evaluated by EQ-5D-3 L questionnaire
Li et al. Trials (2015) 16:583 Page 6 of 12
7. Oxford Hip and Knee Scores (where infection is in standard in studies of medication where adherence is
the hip or knee) critical [19, 20]. Sensors in the caps detect opening and
8. Adherence to oral medication. closing, and record these events with a date stamp. The
sensors are read at a later date to verify whether patients
The study clinicians determine secondary endpoints 1, opened and closed their bottles at times that are consist-
2, 4 and 5. The blinded endpoint review committee ent with their prescription. MEMS are used only with
determines primary endpoints and secondary endpoint specific consent from participants. If more than one
3, by using redacted notes. Participant questionnaires antibiotic is prescribed, we use the MEMS sensors on
determine secondary endpoints 6 and 7. Secondary end- the more frequently dosed antibiotic. If changes to
point 8 is determined by questionnaire in all centres, antibiotic prescriptions are required after discharge,
and by medication event monitoring systems (MEMS) at trial participants continue with only paper adherence
four sentinel sites. questionnaires.
d) Presence of deep pus close to but not adjacent to treatment failure that persist beyond the end of the trial.
bone/prosthetic joint/orthopaedic device Taking these estimates together, we will extrapolate
e) Presence of peri-prosthetic necrotic bone the costs beyond the period of observation within the
f ) Rapid loosening of a joint prosthesis/orthopaedic year of follow-up in the trial. This will necessarily in-
device (that is, leading to localised pain in less than volve a series of assumptions in applying estimates
3 months since implantation) in the absence of a from the literature, and extensive sensitivity analyses
mechanical explanation for rapid loosening. will be examined in order to explore the robustness
of the estimates.
Infection is categorised as “possible” where microbio-
logical sampling has been undertaken with negative re- Trial management and quality assurance procedures
sults (according to criteria described above for “definite” The study is conducted in accordance with the
infection) plus other criteria for definite infection are current approved protocol, International Conference
not fulfilled and, in addition, one or more of the criteria on Harmonisation (ICH) Good Clinical Practice
listed a) to f ) above is met. The review committee mem- guideline, relevant regulations and standard operating
bers are blinded to treatment allocation and subsequent procedures, including data protection.
outcome. Secondary analysis will evaluate non-inferiority We undertake remote monitoring of data entered in
for “definite” or “definite”/“probable” infections only. real time on a secure, anonymised database and conduct
monitoring visits of collaborator sites to confirm integ-
Health economic analysis rity of data.
The health economic evaluation has two parts. The first,
a within-trial analysis, will be performed based on the Data monitoring committee
resource use and health-related quality of life (EQ-5D- The DMC is composed of three members, two of whom
3 L) data. We will use the British National Formulary for are specialists in infectious diseases and the other a
antibiotic costs (with a sensitivity analysis for hospital senior statistician. None are involved in recruitment,
pharmacy discounts). We will include the costs associ- randomisation or follow-up for trial participants or con-
ated with IV administration based on staffing require- tribute to the trial in any way other than through the
ments, equipment cost, clinic visits and transport costs DMC. The DMC met to discuss the study design and
for patient visits as observed in the trial. For unplanned standard operating procedures shortly before the start of
inpatient stays and additional outpatient attendances the study. The DMC evaluate the frequency of endpoints
other than those related to IV administration, we will in an unblinded analysis annually in a closed meeting
use standard NHS reference costs. and make recommendations to the trial steering com-
We will calculate mean costs in each arm of the trial mittee. Although the DMC may recommend suspension
and differences in costs between the two arms, with or cessation of the trial at any time, it is expected
95 % confidence intervals. The EQ-5D-3 L instrument that they would only recommend early stopping if
will be used to estimate per-patient quality-adjusted life there was a very significantly worse outcome in the
years (QALY) with adjustment for any differences PO antibiotic group as compared to the IV group, as
between the groups in EQ-5D-3 L at baseline. Non-para- determined for example by the Haybittle-Peto stop-
metric bootstrapping techniques will be employed to ping boundary.
confirm the robustness of the statistical analysis of
cost, QALY and cost per QALY. Uncertainty in cost- Trial steering committee
effectiveness will be represented on the cost-effectiveness The trial steering committee (TSC) consists of two inde-
plane and as confidence intervals for cost-effectiveness pendent co-chairs (Graham Cook, Imperial College
ratios, or as cost-effectiveness acceptability curves, as London and John Paul, Health Protection England), two
appropriate. public/patient group representatives (Fraser Old, Nuffield
The second part of the analysis will be to extrapolate Orthopaedic Centre Network and Jennifer Bostock,
the observed results in OVIVA beyond the clinical Healthcare-Associated Infection Service Users Research
trial, in order to explore the potential lifetime cost- Forum), and the chief investigator, supported by the trial
effectiveness of a switch in antibiotic administration physician, coordinator and statistician.
route strategy. This extrapolation will be made in The TSC met first at the start of the trial, and then
each diagnostic group, using estimates of long-term meets yearly to review recruitment rates, protocol
recurrence from the literature and the observed re- amendments, and any protocol deviations identified.
currence rates observed within the period of the trial. The co-chairs of the TSC receive recommendations dir-
We will also use the published longer term costs associ- ectly from the DMC and may make recommendations
ated with disability in order to reflect the consequences of to the sponsor regarding the running of the trial.
Li et al. Trials (2015) 16:583 Page 9 of 12
this trial are likely to benefit patients, the NHS and the previous infections or the extent of surgical debride-
health economy. ment. Thirdly, the randomisation process should ensure
A further potential benefit of targeted oral therapy that heterogeneity will be matched across the two arms.
includes a reduction in use of broad-spectrum antibi- Where possible, subgroup analyses will aim to identify
otics and the consequent risk for emerging antibiotic any significant differential effect within defined popula-
resistance. tions. Because there are numerous recruiting centres,
Finally, there is a clear mandate from the Department ranging from specialist units to district general hospitals,
of Health to ensure patient-centred treatment including, the data collected should be an accurate representation
where possible, limitation of hospital attendances, pro- of patients and practice across the NHS.
motion of an independent ”normal life style” and greater Eligibility for recruitment to the trial is based upon
patient choice over their own treatment [30]. clinical criteria rather than diagnostic laboratory results.
There are some limitations associated with the trial. There are several reasons why we have not included
The OVIVA trial is an open label study. The decision histological or microbiological results as part of the in-
to use this design was based on two principles. Firstly, clusion criteria. Firstly, between 12–28 % of bone and
exposure of patients to a placebo IV therapy for a period joint infections diagnosed clinically are not confirmed in
of up to 6 weeks would pose unnecessary risks associ- the laboratory [31, 32], as a result, for example, of prior
ated with the use of an intravascular access device and exposure to antibiotics or sampling error. Nonetheless,
would therefore be unethical. Secondly, due to the num- these patients are treated as infection based on clinical
ber of different antibiotics required to provide optimal criteria. Secondly, the results of laboratory tests, particu-
care for all patients randomised, it was not feasible to larly the histology results, are not always complete
provide matched placebos in every case. Although an within seven days of sampling; had we relied upon la-
open label design leaves the trial open to bias, almost boratory results as part of the inclusion criteria, many
certainly in favour of IV therapy, the endpoints are de- patients would have had to be excluded from this trial
termined according to predefined criteria by an inde- on account of this delay. Thirdly, the pragmatic design
pendent committee who are blinded to treatment of this trial gives due authority over clinical management
allocation. This is achieved through redaction from case to the surgeon or physician responsible for the patient.
notes of any information that might betray the treatment If, according to a research definition, infection was
allocation (for example, reference to IV access devices, deemed not present, the trial could potentially under-
OPAT, drug names, therapeutic drug monitoring). Pri- mine a clinician’s autonomy to treat an infection based
mary endpoints are defined by objective clinical and on clinical criteria alone. Finally, in order to account for
microbiological criteria, assessment of which requires the possibility that uninfected patients are included,
attendance at, or admission to, hospital. They are there- every case which fails to meet a strict prospective defin-
fore hard endpoints, the interpretation of which is un- ition of infection is reviewed by an independent commit-
likely to be influenced by treatment allocation or other tee for a consensus decision on their infection status at
confounding factors. the time of recruitment. The results from these reviews
OVIVA is an inclusive study; there is no selection by will be reflected in the presentation of results.
organism, procedure or surgical site. We recognise that There are two circumstances in which an apparent de-
this results in a heterogeneous study population, but be- viation from allocated treatment arm might arise. Firstly,
lieve that the advantages of generalisability outweigh the for participants randomised to IV therapy, the use of
disadvantages. A more selective recruitment strategy, purely adjunctive oral agents such as rifampicin is allow-
such as inclusion only of primary arthroplasty infections, able. This may at first seem counterintuitive in a study
would have eliminated concern around heterogeneity. which aims to compare IV with oral therapy but is based
However, with an eligible annual population of around upon common practice outside the context of the trial.
1,500 in the UK, recruitment to such a trial would have The principle stems from the fact that, for some oral
been prohibitively long, and the question would remain agents, bioavailability is close to 100 %. There is there-
unanswered for many of the circumstances under which fore no scientific rationale to suggest any advantage of
we manage bone and joint infection in real life. Sec- IV therapy for these agents. Examples include oral ri-
ondly, the hypothesis of the trial is based strictly upon fampicin, which is routinely used alongside IV therapy
the pharmacokinetic principle that appropriately selected in the management of biofilm-associated staphylococcal
oral antibiotics can provide similar tissue concentrations disease, and metronidazole, which is commonly used in
as compared to intravenous antibiotics. This principle is polymicrobial osteomyelitis. To exclude patients allo-
highly unlikely to be influenced by a differential effect cated to the IV arm but who require adjunctive oral
determined by, for example, the site of infection, the therapy would likely incur a bias in favour of oral ther-
presence or absence of metalwork, the number of apy. Secondly, participants randomised to the oral arm
Li et al. Trials (2015) 16:583 Page 11 of 12
short course of IV therapy in a small minority of pa- Additional file 2: OVIVA Consent Form. Form used in the OVIVA trial
to record informed consent. (PDF 13 kb)
tients is unlikely to significantly influence the results. To
ensure transparency around both of these circumstances,
Abbreviations
all antibiotic use (including dose, route of administration CI: chief investigator; CRF: case report form; DMC: data monitoring committee;
and duration) will be recorded from the day of random- ERC: endpoint review committee; GCP: good clinical practice; GP: general
isation through to one year follow-up, and appropriate practitioner; HTA: Health Technology Assessment; ICH: International Conference
on Harmonisation; IV: intravenous; MEMS: medication events monitoring
analyses will be presented. system; NHS: National Health Service; NIHR: National Institute for Health
Initial recruitment of study patients was slower than Research; OHS: Oxford Hip Score; OKS: Oxford Knee Score; OPAT: outpatient
anticipated. This was primarily due to delays in site parenteral antimicrobial therapy; PIS: Patient Information Sheet; PO: per oral;
PROM: patient reported outcome measure; QALY: quality-adjusted life years;
setup and in establishing mechanisms for the timely REC: Research Ethics Committee; SAE: serious adverse event; SMPC: summary
identification and referral of potential participants to of product characteristics; TSC: trial steering committee.
the research teams. Recruitment was also adversely
Competing interests
affected by a surprising number of patients who, hav- The authors declare that they have no competing interests.
ing read the PIS and understanding the principle of
equipoise, requested oral therapy rather than IV ther- Authors’ contributions
The research question was identified by PB with significant contributions
apy. In order to achieve the recruitment target, we from MS, ASW, BAn, BA, AW, IB, AB, GT, HP, AS and BAL. HKL, MS, RZ, CC and
implemented a web-based network site, monthly tele- PB established the trial operationally. ASW and IR provided statistical trial
conferences and recruitment drives, and requested an design expertise and ABr expertise in health economic analysis. Additional
specialist clinical input into the protocol design and manuscript preparation
extension without additional cost to the funding was provided by MM, DS and BAL. HKL, MS, CC, IR and RZ were responsible
agency. for the initial draft of the manuscript. The views and opinions expressed in
At the second planned interim analysis, the overall pri- this publication are those of the authors and do not necessarily reflect those
of the National Health Service or the NIHR. All authors read and approved
mary endpoint event rate (12.5 %) was significantly the final manuscript.
higher than had been predicted from the single centre
pilot study (5 %); reasons for this might include stochas- Acknowledgements
tic variation related to a small sample, the effect of re- The OVIVA study is funded by the National Institute for Health Research
(Health Technology Assessment project number 11/36/29). The Nuffield
cruitment at a single specialist centre or a combination Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences
of factors. In order to accommodate this unexpected (NDORMS) coordinates the study via the Surgical Interventions Trials Unit
finding, and with appropriate ethical permission and en- (SITU). The trial sponsor is Oxford University Hospitals NHS Trust.
The authors greatly appreciate the considerable contributions of Deepa Bose,
dorsement from the DMC, we elected to amend the Harriet Hughes, Neil French, Martin Llewelyn, Colette Smith and David Beard
non-inferiority margin from 5 % to 7.5 %. This figure re- in refining the trial design, acquisition and interpretation of data.
mains within published guidelines, which suggest that a We acknowledge the considerable support of Jennifer Bostock, Fraser Old
(public patient representatives), Alice Harin, Lorrayne Jefferies, Louise Spoors,
non-inferiority margin of 10 % is appropriate for most Patrick Julier and all members of the infectious diseases and orthopaedic
therapeutic intervention trials in infection [33]. We departments at Oxford University Hospitals NHS Trust. We thank members of
therefore believe that the non-inferiority margin remains the United Kingdom Clinical Infection Research Group (UKCIRG) for their
continued involvement.
appropriate for this trial and is unlikely to jeopardise the We are extremely grateful to the funding agency and to all patients who
utility of the results. have participated.
Despite its limitations, the trial will be the largest
Author details
study of its type addressing this question of oral versus 1
Nuffield Department of Medicine, University of Oxford, Oxford, UK. 2Oxford
IV antibiotic therapy in bone and joint infection, and is University Hospitals NHS Trust, Oxford, UK. 3Nuffield Department of
likely to have important implications for patients and Orthopaedics, Rheumatology and Musculoskeletal Science, University of
Oxford, Oxford, UK. 4MRC Clinical Trials Unit, University College London,
healthcare practitioners in the field of orthopaedic infec- London, UK. 5Division of Medical Sciences, University of Oxford, Oxford, UK.
tion, and for the health economy. 6
Health Economics and Health Technology Assessment, University of
Li et al. Trials (2015) 16:583 Page 12 of 12
Glasgow, Glasgow, UK. 7Gartnavel General Hospital, NHS Greater Glasgow 20. Farley J, Hines S, Musk A, Ferrus S, Tepper V. Assessment of adherence to
and Clyde, Glasgow, UK. 8Heart of England NHS Foundation Trust, antiviral therapy in HIV-infected children using the Medication Event
Birmingham, UK. 9Oxford University Clinical Research Unit, Wellcome Trust, Monitoring System, pharmacy refill, provider assessment, caregiver
Ho Chi Minh, Vietnam. 10Kenya Medical Research Institute, Wellcome Trust, self-report, and appointment keeping. J Acquir Immune Defic Syndr.
Kilifi, Kenya. 2003;33:211–8.
21. Waldvogel FA, Medoff G, Swartz MN. Osteomyelitis: a review of clinical
Received: 21 August 2015 Accepted: 2 December 2015 features, therapeutic considerations and unusual aspects. N Engl J Med.
1970;282(4):198–206.
22. Berendt A, McNally M. Osteomyelitis. In: Warrell DA, Cox TM, Firth JD,
editors. Oxford textbook of medicine. 5th ed. Oxford: Oxford University
References Press; 2010. p. 3788.
1. National Joint Registry. NJR 10th Annual Report 2013; ISSN 2054-183X. 23. Mader JT, Calhoun J. Osteomyelitis. In: Mandell GL, Bennett JE, Dolin R,
http://www.njrcentre.org.uk/njrcentre/Portals/0/Documents/England/Reports/ editors. Principles and practice of infectious diseases. 4th ed. London:
10th_annual_report/NJR%2010th%20Annual%20Report%202013%20B.p. Churchill Livingstone; 1995. p. 1039.
Accessed 10 March 2015. 24. Osmon DR, Berbari EF, Berendt AR, Lew D, Zimmerli W, Steckelberg JM, et
2. Public Health England. Surveillance of surgical site infections in NHS hospitals al. Diagnosis and management of prosthetic joint infection: clinical practice
in England, 2013/2014. London: Public Health England. https://www.gov.uk/ guidelines by the Infectious Diseases Society of America. Clin Infect Dis.
government/uploads/system/uploads/attachment_data/file/386927/SSI_ 2013;56(1):e1–25.
report_2013_14_final__3_.pdf. Accessed 10 March 2015. 25. Fraimow HS. Systemic antimicrobial therapy in osteomyelitis. Semin Plast
3. Klouche S, Sariali E, Mamoudy P. Total hip arthroplasty revision due to Surg. 2009;23(2):90–9.
infection: a cost analysis approach. Orthop Traumatol Surg Res. 26. Johannsson B, Taylor J, Clark CR, Shamsuddin H, Beekmann SE, Polgreen P,
2010;96(2):124–32. et al. Treatment approaches to prosthetic joint infections: results of an
4. Vanhegan IS, Malik AK, Jayakumar P, Ul Islam S, Haddad FS. A financial Emerging Infections Network survey. Diagn Microbiol Infect Dis.
analysis of revision hip arthroplasty: the economic burden in relation to 2010;66(1):16–23.
the national tariff. J Bone Joint Surg (Br). 2012;94(5):619–23. 27. Stengel D, Bauwens K, Sehouli J, Ekkernkamp A, Porzsolt F. Systematic
5. Kallala RF, Vanhegan IS, Ibrahim MS, Sarmah S, Haddad FS. Financial analysis review and meta-analysis of antibiotic therapy for bone and joint infections.
of revision knee surgery based on NHS tariffs and hospital costs: does it pay Lancet Infect Dis. 2001;1(3):175–88.
to provide a revision service? Bone Joint J. 2015;97-B(2):197–201. 28. Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB.
6. Lew DP, Waldvogel FA. Osteomyelitis. Lancet. 2004;364:369–79. Nosocomial bloodstream infections in US hospitals: analysis of 24,179
7. Zimmerli W, Ochsner PE. Management of infection associated with cases from a prospective nationwide surveillance study. Clin Infect Dis.
prosthetic joints. Infection. 2003;31:99–108. 2004;39(3):309–17.
8. Legrand E, Flipo RM, Guggenbuhl P, Masson C, Maillefert JF, Soubrier M et al. 29. Gupta A, Swaroop C, Agarwala S, Pandey RM, Bakhshi S. Randomised
Management of nontuberculous infectious discitis. Treatments used in 110 controlled trial comparing oral amoxicillin-clavulanate and ofloxacin with
patients admitted to 12 teaching hospitals in France. Joint Bone Spine. intravenous ceftriaxone and amikacin as outpatient therapy in pediatric
2001;68(6):504–9. low-risk febrile neutropenia. J Pediatr Hematol Oncol. 2009;31(9):635–41.
9. Oosterheert JJ, Bonten MJ, Schneider MM, Buskens E, Lammers JW, Hustinx 30. Hackett M. Homecare Medicines “Towards a vision for the future”.
WM et al. Effectiveness of early switch from intravenous to oral antibiotics Department of Health Best Practice Guidance; 2011. Ref 16691. https://
in severe community acquired pneumonia: multicentre randomised trial. www.gov.uk/government/uploads/system/uploads/attachment_data/file/
BMJ. 2006;333(7580):1193. 213112/111201-Homecare-Medicines-Towards-a-Vision-forthe-Future2.pdf.
10. Montini G, Toffolo A, Zucchetta P, Dall’Amico R, Gobber D, Calderan A, et al. 10 March 2015.
Antibiotic treatment for pyelonephritis in children: multicentre randomised 31. Berbari EF, Hanssen AD, Duffy MC, Steckelberg JM, Ilstrup DM, Harmsen WS,
controlled non-inferiority trial. BMJ. 2007;335:386. et al. Risk factors for prosthetic joint infection: case-control study. Clin Infect
11. Shenep JL, Flynn PM, Baker DK, Hetherington SV, Hudson MM, Hughes WT, Dis. 1998;27(5):1247–54.
et al. Oral cefixime is similar to continued intravenous antibiotics in the 32. Sheehy SH, Atkins BA, Bejon P, Byren I, Wyllie D, Athanasou NA, et al. The
empirical treatment of febrile neutropenic children with cancer. Clin Infect microbiology of chronic osteomyelitis: prevalence of resistance to common
Dis. 2001;32:36–43. empirical anti-microbial regimens. J Infect. 2010;60(5):338–43.
12. Stevens DL, Herr D, Lampiris H, Hunt JL, Batts DH, Hafkin B. Linezolid versus 33. Addendum to the guideline on the evaluation of medicinal products
vancomycin for the treatment of methicillin-resistant Staphylococcus aureus indicated for treatment of bacterial infections. http://www.ema.europa.
infections. Clin Infect Dis. 2002;34:1481–90. eu/docs/en_GB/document_library/Scientific_guideline/2013/11/
13. Heldman AW, Hartert TV, Ray SC, Daoud EG, Kowalski TE, Pompili VJ, et al. WC500153953.pdf. Accessed 15 July 2015
Oral antibiotic treatment of right-sided staphylococcal endocarditis in
injection drug users: prospective randomized comparison with parenteral
therapy. Am J Med. 1996;101:68–76.
14. Conterno LO, da Silva Filho CR. Antibiotics for treating chronic osteomyelitis
in adults. Cochrane Database Syst Rev. 2009; 8(3):CD004439.
15. Whittaker JP, Warren RE, Jones RS, Gregson PA. Is prolonged systemic
antibiotic treatment essential in two-stage revision hip replacement for
chronic Gram-positive infection? J Bone Joint Surg (Br). 2009;91:44–51.
16. Hart WJ, Jones RS. Two-stage revision of infected total knee replacements Submit your next manuscript to BioMed Central
using articulating cement spacers and short-term antibiotic therapy. J Bone and we will help you at every step:
Joint Surg (Br). 2006;88:1011–5.
17. Matthews PC, Conlon CP, Berendt AR, Kayley J, Jefferies L, Atkins BL, et al. • We accept pre-submission inquiries
Outpatient parenteral antimicrobial therapy (OPAT): is it safe for selected • Our selector tool helps you to find the most relevant journal
patients to self-administer at home? A retrospective analysis of a large
• We provide round the clock customer support
cohort over 13 years. J Antimicrob Chemother. 2007;60:356–62.
18. Ceroni D, Regusci M, Pazos JM, Saunders CT, Kaelin A. Risks and • Convenient online submission
complications of prolonged parenteral antibiotic treatment in children with • Thorough peer review
acute osteoarticular infections. Acta Orthop Belg. 2003;69(5):400–4.
• Inclusion in PubMed and all major indexing services
19. Diaz E, Levine HB, Sullivan MC, Sernyak MJ, Hawkins KA, Cramer JA, et al.
Use of the Medication Event Monitoring System to estimate medication • Maximum visibility for your research
compliance in patients with schizophrenia. J Psychiatry Neurosci.
2001;26(4):325–9. Submit your manuscript at
www.biomedcentral.com/submit